If progressive politicians and journalists want to advocate for redistribution of income via 70% tax rates on high income earners, then just do it. But don’t use the argument that “income tax rates on the rich used to be much higher in the past” as justification for it, since the top 1% paid around the same amount relative to income back then as they do now.

Furthermore, if you want to argue for wealth taxes on households with assets over $50 mm (as Elizabeth Warren now proposes), then just do it. But if you are going to base the rationale for such taxes on work done by Saez/Piketty/Zucman, be aware of limitations in their work, and pointed questions about their assumptions and methodology from the US Treasury, the Federal Reserve, the Joint Committee on Taxation and professors at Columbia University.

By now I’m sure you’ve read articles on proposals for a 70% tax rate on income above certain thresholds. Progressive politicians and journalists have every right to advocate for such taxes as a way to redistribute income (to be clear, the US Federal income tax is already progressive, and remains so after adjusting for state and local taxes1). But when people base such measures on the notion that “we’re just returning to fairer tax rates that existed in the past”, this is pretty sloppy work, which is ironic since a lot of these individuals/articles blame others for distorting facts.

Yes, statutory tax rates on the top 1% of households used to be much higher. But this fact is almost useless without understanding what the rest of the tax code entailed regarding bracket levels2 and deductions. The CBO computes “effective tax rates”, which is a pretty simple concept: total taxes paid as a percentage of income. As shown in the first chart, when the top Federal statutory rate was 70% in 1979, effective income tax rates on the top 1% were roughly the same as they are today. Even Saez and Piketty (more on them later) show that effective tax rates on the top 1% were not much higher in the 1950’s when including all Federal, state and local income, payroll, property, excise and other taxes (second chart). Bottom line: history does not provide apples-to-apples precedent for 70% marginal tax rates, so people should stop saying it does3. Base your arguments on the future rather than on the past.

A chart using two lines to compare the top statutory federal income tax rate to the effective federal tax rate for the top 1% from 1979 through 2015. In all years, the effective tax rate has been significantly lower than the statutory rate.

Line chart tracking the effective tax rate, including all federal, state and local taxes, on the top 1% from 1955 to 2015.

Wealth taxes. Senator Elizabeth Warren4 has announced a plan for a 2% annual “wealth tax” on household assets in excess of $50 mm (see Appendix on wealth tax constitutionality and possible de facto alternatives). As per a January Washington Post article, Warren’s plan is based on work done by Saez, Piketty and Zucman, some of whom are also advising Senator Warren directly. We have written about their work before, in September 20175. It turns out that wealth shares by income level are not easy to obtain, and can only be estimated using a lot of assumptions.

Economists at the Federal Reserve who also studied US wealth distribution issued a stark rebuttal of the Saez/Piketty analysis, expressing concern that little attention has been paid to their methodology or to the sensitivity of their results to changing assumptions6, and concluded that Saez’ methods can yield “improbable” results. Alternative approaches to deriving 1% wealth shares shown in the first chart also indicate high concentrations of wealth, but they may not have changed much at all since the 1960s, in contrast to the Saez/Piketty analysis.

Researchers have also taken another look at Saez’ work on income inequality. In another rebuttal, economists from the US Treasury and the Joint Committee on Taxation present a very different picture of income inequality than widely-cited Saez/Piketty results7, as shown in the 2nd chart. Treasury/JCT authors claim a more accurate measurement of income; remove distortions from changing tax laws affecting the number of filers; and expand the definition of income. They conclude by citing an increase in income inequality since 1960 that is just one tenth of Saez/Piketty findings. Here’s how the Treasury/JCT paper describes Saez’ use of unadjusted tax returns to derive income levels: “These results show that unadjusted tax return based measures present a distorted view of inequality trends, as incomes reported on tax returns are sensitive to changes in tax laws and ignore income sources outside the individual tax system.”

A three-line graph tracking the share of national wealth owned by the top 1% as measured by the Capitalization method, the Survey of Consumer Finances method, and the estate tax multiplier method.

A two-line chart showing the percentage of total national income earned by the top 1% from 1960 to 2015 as measured by both the Saez/Piketty and Treasury/JCT methods.
Perhaps one of the most corrosive things the current Administration has done is to lower the bar for false or misleading statements. Given the first chart below8, it may be difficult to get people to care very much right now about exaggerations, academic bias and misleading statements on tax related issues. In any case, neither party has an unblemished record for straight talk9, and as illustrated in the second chart using empirically derived data10, we may be headed into the most ideologically charged Presidential election campaign in 100 years. Bottom line: don’t believe everything you read.

A line chart showing the rise in the cumulative number of false or misleading claims made by President Trump from January 2017 to January of 2019, per the Washington Post. The number exceeds 8,000.

A line chart that rates the degree of conservative or liberal leanings of administrations from 1924 to the present day, and that evaluates the degree of progressivism of leading Democratic candidates seeking the 2020 presidential nomination.

Appendix: On the constitutionality of wealth taxes and de facto alternatives

The first reaction many people have to idea of wealth taxes: “they’re probably unconstitutional”, since the 16th Amendment only sanctions income taxes, and prior Constitutional provisions preclude taxes on the states that are not apportioned back in the same ratio. However, some tax scholars have already identified de facto alternatives to a wealth tax using Wealth Integration Methods. The simple explanation: instead of taxing wealth directly, wealth is used as an input into income tax calculations to get to roughly the same place. This approach might seem just like a Wealth Tax, but a recent paper from a law professor at Ohio State discounts possible Constitutional objections to Wealth Integration Taxes, arguing that they are not precluded by the 16th Amendment, and that to reject them would require the Courts to overturn settled precedent11. The author goes so far as to say that since the Courts are likely to accept Wealth Integration Taxation methods, they might as well just allow Congress to tax wealth directly. Welcome to 2019. Likely candidates for a Wealth Integration Tax: deduction phase-outs based on wealth, and/or increasing marginal tax rates based on wealth levels. Furthermore, the definition of wealth could require liquid assets to be valued on a mark-to-market basis, subjecting taxpayers to de facto wealth taxes whether or not assets are sold; and for these households, income and capital gains tax rates will probably be unified at the same level.